A new simulation of distributed mutual exclusion on neural networks

In a distributed system, process synchronization is an important agenda. One of the major duties for process synchronization is mutual exclusion. In new algorithm, opposite the past algorithms fairness happens. This paper presents a new approach of the race models involving distributed mutual exclusion. Further, concrete applications of these models did not involve variability in the accumulator size or were based on a specific distribution. We show that the distributions of time stamp, time action and the other effective parameters predicted by the neural network competitive models can be solved analytically this problem that be happens in the critical sections. The model can be manipulated and simulated to predict the effects of reward on Hamming and Hopfieldpsilas models curves and speed-accuracy decomposition. In other hand, the major contribution of this paper is the implementation of a learning rule that enables networks based on a race model to learn stimulus-response associations. The model described here can be seen as a reduction of information system and is compatible with a priority learning system. Also, we will consider the non-linear behavior of the competitive models and as a result use this property in distributed systems. Finally, it is possible to use the neural networks as a distributed system pattern, to optimization of fault tolerance, reliability and accessibility related to mutual exclusion and critical section. Thus in the new approach fault tolerance will ascend and centralize and distributed algorithms can use this and based algorithm will be more reliable.